Alzheimer's disease (AD) is a devastating neurodegenerative progressive disorder, which is the predominant cause of dementia in people over 65 years of age. The prevalence of AD is estimated to be as high as 18.7% among 75-84 year-olds and 47.2% among the .gtoreq.85 year age groups, affecting a significant portion of the population in most countries of the world.
Clinical symptoms of the disease typically begin with subtle short term memory problems. As the disease progresses, difficulty with memory, language, and orientation worsen to the point of interfering with the ability of the person to function independently. Other symptoms, which are variable, include myoclonus and seizures. Duration of AD from the first symptoms of memory loss until death is 10 years on average, but may range from 6-8 years to more than 20 years. AD always results in death, often from respiratory-related illness.
The pathology in AD is confined exclusively to the central nervous system (CNS). The AD brain is characterized by the presence of amyloid deposits and neurofibrillary tangles (NFT).
Amyloid deposits are found associated with the vascular system of the CNS and as focal deposits in the parenchyma. The major molecular component of an amyloid deposit is a highly hydrophobic peptide called the A.beta. peptide. This peptide aggregates into filaments in an anti-.beta.-pleated sheet structure resulting in the birefringent nature of the AD amyloid. Although A.beta. is the major component of AD amyloid, other proteins have also found associated with the amyloid, e.g., .alpha.-1-anti-chymotrypsin (Abraham, et al., Cell 52:487-501 (1988)), cathepsin D (Cataldo et al., Brain Res. 513:181-192 (1990)), non-amyloid component protein (Ueda et al., Proc. Natl. Acad. Sci. USA 90:11282-11286 (1993)), apolipoprotein E (apoE) (Namba et al., Brain Res. 541:163-166 (1991); Wisniewski & Frangione, Neurosci. Lett. 135:235-238 (1992); Strittmatter et al., Proc. Natl. Acad. Sci. USA 90:1977-1981 (1993)), apolipoprotein J (Choi-Mura et al., Acta Neuropathol. 83:260-264 (1992); McGeer et al., Brain Res. 579:337-341 (1992)), heat shock protein 70 (Hamos et al., Neurology 41:345-350 (1991)), complement components (McGeer & Rogers, Neurology 43:447-449 (1992)), .alpha..sub.2 -macroglobin (Strauss et al., Lab. Invest. 66:223-230 (1992)), interleukin-6 (Strauss et al., Lab. Invest. 66:223-230 (1992)), proteoglycans (Snow et al., Lab. Invest. 58:454-458 (1987)), and serum amyloid P (Coria et al., Lab. Invest. 58:454-458 (1988)).
Plaques are often surrounded by astrocytes and activated microglial cells expressing immune-related proteins, such as the MHC class II glycoproteins HLA-DR, HLA-DP, and HLA-DQ, as well as MHC class I glycoproteins, interleukin-2 (IL-2) receptors, and IL-1. Also surrounding many plaques are dystrophic neurites, which are nerve endings containing abnormal filamentous structures.
The characteristic Alzheimer's NFTs consist of abnormal filaments bundled together in neuronal cell bodies. "Ghost" NFTs are also observed in AD brains, which presumably mark the location of dead neurons. Other neuropathological features include granulovacular changes, neuronal loss, gliosis and the variable presence of Lewy bodies.
The destructive process of the disease is evident on a gross level in the AD brain to the extent that in late-stage AD, ventricular enlargement and shrinkage of the brain can be observed by magnetic resonance imaging. The cells remaining at autopsy, however, are grossly different from those of a normal brain, characterized by extensive gliosis and neuronal loss. Neurons which were possibly involved in initiating events, are absent; and other cell types, such as the activated microglial cells and astrocytes, have gene expression patterns not observed in the normal brain. Thus, the amyloid plaque structures and NFTs observed at autopsy are most likely the end-products of a lengthy disease process, far removed from the initiating events of AD.
Accordingly, attempts to use biochemical methods to identify key proteins and genes in the initiating steps of the disease are hampered by the fact that it is not possible to actually observe these critical initiating events. Rather, biochemical dissection of the AD brain at autopsy is akin to molecular archeology, attempting to reconstruct the pathogenic pathway by comparing the normal brain to the end-stage disease brain.
Substantial evidence has suggested that inherited genetic defects are involved in AD. Numerous kindreds have been described in the literature as having early-onset AD (defined as onset before age 65). Bird et al., Ann. Neurol. 23:25-31 (1988); Bird et al., Ann. Neurol, 25:12-25 (1989); Cook et al., Neurology 29:1402-1412 (1979); Feldman et al., Neurology 13:811-824 (1960); Goudsmit, J. Neurol. Sci. 49:79 (1981); Heston & White, Behavior Genet. 8:315-331 (1978); Martin et al., Neurology 41:62-68 (1991); Nee et al., Arch. Neurol. 40:203-208 (1983); van Bogaeert et al., Mschr. Psych. Neurol. 102:249-301 (1940); Wheelan, Ann. Hum. Genet. 23:300-309 (1959)). Families with multiple late-onset AD cases have also been described (Bird et al., Ann. Neurol. (1989), supra; Heston & White, Behavior Genet. (1978), supra; Pericak-Vance et al., Exp. Neurol. 102:271-279 (1988)). In addition, twin studies have documented that monozygotic twins are more concordant in their AD phenotype than dizygotic twins (Nee et al., Neurology 37:359-363 (1987). Also, the families of concordant twins have more secondary cases of AD than families of discordant twins (Rapoport et al., Neurology 41:1549-1553 (1991)).
Genetic dissection of AD has been complicated by the complexity and overall accuracy of its diagnosis. Because AD is relatively common in the elderly, clustering of cases in a family may occur by chance, representing possible confounding non-allelic genetic heterogeneity, or etiologic heterogeneity with genetic and non-genetic cases co-existing in the same kindred. In addition, the clinical diagnosis of AD is confounded with other dementing diseases common in the elderly.
Despite these problems, mutations in the amyloid precursor protein (APP) gene on chromosome 21 have been associated with early-onset (&lt;65 years) autosomal dominant AD (Goate et al., Nature 349:704 (1991)). Moreover, mutations in two recently identified genes, S182 on chromosome 14 and STM-2 on chromosome 1, which encode presenilin 1 (PS1) and presenilin 2 (PS2), respectively, have also been associated with early-onset autosomal dominant AD (Schellenberg et al., Science 258:668 (1992); Sherrington et al., Nature 375:754-760 (1995); Levy-Lahad/Wasco et al., Science 269:973-977 (1995)).
For late-onset AD, the APOE gene has been identified as a genetic modifying factor (Strittmatter et al., Proc. Natl. Acad. Sci. USA 90:1977 (1993); Corder et al., Science 261:921 (1993); Corder et al., Nat. Genet. 7:180-184 (1994); Benjamin et al., Lancet 344:473 (1994); Smith et al., Lancet 344:473-474 (1994)).
However, the known genetic loci for AD do not account for all cases of AD. For example, in late-onset AD approximately half of AD cases do not have the APOE .epsilon.4 allele found in several other families with high incidence of AD, including the Volga German (VG) kindreds. Brousseau et al., Neurology 342 (1994); Kuusisto et al., Brit. Med. J. 309:363 (1994); Tsai et al., Am. J. Hum. Genet. 54:643 (1994); Liddel et al., J. Med. Genet. 31:197 (1994); Cook et al., Neurology (1979), supra; Bird et al., Ann. Neurol. (1988), supra; Bird et al., Ann. Neurol. 25:12 (1989). The known AD loci have been excluded as possible causes of the discrepancy. Schellenberg et al., Science (1992), supra; Lannfelt et al., Nat. Genet. 4:218-219 (1993)); van Duijn et al., Am. J Hum. Genet. 55:714-727 (1994); Schellenberg et al., Science 241:1507 (1988); Schellenberg et al., Am. J. Hum. Genet. 48:563 (1991); Schellenberg et al., Am. J. Hum. Genet. 49:511-517 (1991); Kamino et al., Am. J. Hum. Genet. 51:998 (1992); Schellenberg et al., Am J. Hum. Genet. 53:619 (1993); Schellenberg et al., Ann. Neurol. 31:223 (1992); Yu et al., Am. Hum. Genet. 54:631 (1994)). Thus, identification of new genes and of risk-modifying alterations of existing genes will add considerably to an understanding of the genetic determinants of AD, and enable biochemical and genetic approaches to the diagnosis and therapeutic treatment.
The present invention provides novel, previously unidentified and apparently pathogenic mutations of the chromosomal loci for familial AD (FAD), methods and compositions for diagnosis and treatment of AD, and other related advantages.